If you're looking for an interesting holiday read, you could do worse that Merrill R. Chapman's "In Search Of Stupidity : Over 20 Years Of High-Tech Marketing Disasters". The premise of the book is that most of the old software companies that used to be around - MicroPro, Ashton-Tate, Borland, Netscape and so on - lost their market share due to stupid marketing rather than the innate superiority of the competition. The foreword by Joel Spolsky sums it up well:

"... If you want to be successful in the software business, you have to have a management team that thoroughly understands and loves programming, but they have to understand and love business, too. Finding a leader with strong aptitude in both dimensions is difficult, but it’s the only way to avoid making one of those fatal mistakes that Rick catalogs lovingly in this book. So read it, chuckle a bit, and if there’s a stupidhead running your company, get your résumé in shape and start looking for a house in Redmond."

In other words, the reason that companies like Microsoft and Oracle are still around now is because they haven't made the same dumb marketing mistakes that most of the competition fall foul of. To illustrate this, Joel points out how the software market has changed over the last ten years.

"A nice fellow named Jeffrey Tarter used to publish an annual list of the hundred largest personal computer software publishers called the Soft-letter 100. Here’s what the top ten looked like in 1984[1]:

Rank	Company	Annual Revenues
#1	Micropro International	$60,000,000
#2	Microsoft Corp.	$55,000,000
#3	Lotus	$53,000,000
#4	Digital Research	$45,000,000
#5	VisiCorp	$43,000,000
#6	Ashton-Tate	$35,000,000
#7	Peachtree	$21,700,000
#8	MicroFocus	$15,000,000
#9	Software Publishing	$14,000,000
#10	Broderbund	$13,000,000

OK, Microsoft is number 2, but it is one of a handful of companies with roughly similar annual revenues.

Now let’s look at the same list for 2001.

Rank	Company	Annual Revenues
#1	Microsoft Corp.	$23,845,000,000
#2	Adobe	$1,266,378,000
#3	Novell	$1,103,592,000
#4	Intuit	$1,076,000,000
#5	Autodesk	$926,324,000
#6	Symantec	$790,153,000
#7	Network Associates	$745,692,000
#8	Citrix	$479,446,000
#9	Macromedia	$295,997,000
#10	Great Plains	$250,231,000

Whoa. Notice, if you will, that every single company except Microsoft has disappeared from the top ten. Also notice, please, that Microsoft is so much larger than the next largest player, it’s not even funny. Adobe would double in revenues if they could just get Microsoft’s soda pop budget.

The personal computer software market is Microsoft. Microsoft’s revenues, it turns out, make up 69% of the total revenues of all the top 100 companies combined.

This is what we’re talking about, here."

The book itself concentrates mainly on the PC hardware and software market (hence not really Oracle's area of operation) and what makes it particularly interesting is that the author (Merrill Chapman) worked at quite a few of them over the years. Some of the stories are particularly entertaining and I'll share a few of them here.

First up is MicroPro, the company behind the WordStar word-processing application. MicroPro, with WordStar, used to 'own' the PC word-processing market, primarily because WordStar's use of a control-key interface was particularly suited to the needs of touch-typists. According to the book, "WordStar's layout was not mnemonic; instead, in the interests of fast typing, Rubenstein designed the interface so that all cursor movements were performed with the left hand while less common operations fell to the right hand. WordStar users came to swear by this system, and today diehards still retrofit Microsoft Word and other products with add-ins and utilities that resurrect the WordStar keyboard system." WordStar was also the first WYSIWYG word processor, and dominated the DOS word processing market in the same way that Microsoft Word dominates the Windows market.

Although WordStar made MicroPro the biggest PC software company in the world, MicroPro managed to blow it through a process that began with the falling out with, then sacking of, their entire development team, who then went on to produce a WordStar clone called NewWord. Then, realising that they had no way of delivering the next version of WordStar, they then bought another company that had built a Unix clone of WordStar to obtain their developers, who were then set to work on producing the next version of WordStar. However (still with me?) this new team, who worked away from the main MicroPro office and had little supervision, and came up with a completely new design for their version of WordStar - one that bore little resemblance to the traditional WordStar interface, and had a completely new user interface and set of command keys. MicroPro had little choice but to name this new product 'WordStar 2000', and pitch it as the logical upgrade to WordStar 3.3

The problem for MicroPro was that when users took a look at WordStar 2000, they decided it was too different to what they were used to, and waited instead for an upgrade to the 'traditional' version of Wordstar. MicroPro then had to go on selling both versions, desperately trying to come up with a convincing rationale for the existence of the two versions, at one point positioning WordStar 2000 as a desktop publishing tool, at other times as the 'power user' version of WordStar. In the end, instead of MicroPro's sales force spending their time selling WordStar, they in fact spent most of their time trying to explain the differences between the two versions and why you might choose one over the other. In the end, they rehired the old development team and eventually released an upgrade to the original WordStar, but by that point Windows was out and Microsoft Word had taken over the market.

Probably the most entertaining story in the book though is around Ashton-Tate, and the DBase database product. Ashton-Tate, at the beginning of the 1980's was one of the big three PC software vendors, along with Microsoft and Lotus. Ashton-Tate owned the database market, Lotus was spreadsheets, and Microsoft had DOS. DBase was originally known as a product called 'Vulcan', which Ashton-Tate bought and renamed DBase II. DBase was one of the first desktop relational databases, and the DBase file format still lives on in products such as XBase and Microsoft Foxbase. The Ashton-Tate story got interesting when the company founder, George Tate, died and Es Esber, his second-in-command, took over. Esber fancied himself as a business guru and intended to bring a bit of 'professionalism' to the company. Despite having a degree in computer engineering, he wasn't known as particularly technically astute.#

Esber looked at the development community around DBase, and saw this as a missed opportunity for Ashton-Tate; every utility developed by a third party was revenue lost to Ashton-Tate, and every developer working with DBase was taking money out of Ashton-Tate's pockets. This was particularly significant as DBase had a very loyal development and partner community, and it was in fact these developers and companies that added to the DBase functionality that made it the number one database application on the PC platform. Esber then went on a one man crusade, threatening to sue anyone who worked as a developer delivering solutions using Dbase, together with third-party vendors who sold applications that were compatible with Dbase. At one point, Esber stood up at a software developers conference and shouted 'Make my day!' while threatening legal action against anyone who wrote DBase-compatible applications.

Imagine Larry Ellison standing up at Oracleworld and threatening to sue any developer who built applications using Oracle. Or threatening action against any vendor who built applications that ran against Oracle Application Server. The upshot of all this was that the entire DBase development community turned against Ashton-Tate, other companies came in with look-alike products, and as we all now know, Microsoft came in with Access and cleaned up in the market.

The stories go on: IBM with OS/2 and the micro channel architecture, Netscape with Navigator and their five year exit from the market to rewrite the code base from scratch. The message that the book keeps coming back to is: the reason companies like Microsoft have succeeded is that they haven't made the same stupid marketing mistakes that the competition has. The others had the lead, but then blew it with a series of bad business decisions.

The book doesn't really touch on Oracle, as it's about the world of PC hardware and software. Thinking about it though, I wonder what a history of the database market would look like, and how would Oracle's marketing decisions be judged over time? For example, how will history judge the following marketing and technical decisions taken by Oracle?

• The decision to move Apps 11i to entirely-web based and integrated  (probably favourably, contributed the huge success of the e-Business Suite at the expense of best-of-breed vendors)
• The decision to brand Oracle 8 as 'Oracle 8i, the internet database' (again probably favourably, although more through catching the dot.com mood rather than because the internet functionality was required)
• The move of developers from Forms and PL/SQL to Java and JDeveloper (jury's out - for Oracle it's clearly been a benefit, but many outside of Oracle are yet to be convinced that JDeveloper is as productive as Forms)
• The choice of Java and J2EE as the programming environment (much of this comes down to whether J2EE or .Net comes out top, and of course Oracle have got a foot in the .Net camp as well)
• The emphasis on Linux as the platform of choice (inspired choice, or removes much of the credibility through association with Big Iron and Unix vendors such as Sun and HP?)
• The latest move into Grid computing (who knows? it's still early days)
• Raw Iron? The Net Appliance? PowerBrowser? OracleOffice? (we'll let them off these :-))

Clearly for Oracle, the gamble is on high-functionality and premium positioning, with J2EE as the programming 'glue' and Linux as the commodity platform of choice. They've seen off many database and ERP vendors, and now the battle is between Oracle's vision and that put forward by IBM and Microsoft, with history (and the book) suggesting that the vendor that is still here in twenty years time is the one who is less stupid than their competitors.

Anyway, it's an excellent read, a great look at the software companies we grew up with but somehow aren't here anymore, thoroughly recommended and well worth putting in your suitcase and taking on holiday.